This is probably a little late, but I've found that hand-soldering the CC1110 is practically impossible. I fought a similar problem last year for a week before my machine reflowed boards came in. Of 3 prototypes I soldered (2 of which were soldered by a professional assembly tech), only 1 ever talked to the debugger, and it had poor RF performance. All of these boards passed the "buzz out" test by testing ground to the 4 corners of the package where the tiny ground pads are.
I just got another CC1110 board in my hands yesterday, and once again I tried to assemble two boards by hand with a hot-air rework station. Neither will talk to the debugger. Note the board has 4 other QFN parts, and all of them work. I even soldered only the pad first and made sure the pad reflowed (I can watch the part center on just the ground pad), but still no debugger comms.